U.S. patent application number 09/941496 was filed with the patent office on 2002-11-07 for system and methods for generating physician profiles concerning prescription therapy practices.
Invention is credited to Cain, James, Tolle, Jill.
Application Number | 20020165736 09/941496 |
Document ID | / |
Family ID | 26956387 |
Filed Date | 2002-11-07 |
United States Patent
Application |
20020165736 |
Kind Code |
A1 |
Tolle, Jill ; et
al. |
November 7, 2002 |
System and methods for generating physician profiles concerning
prescription therapy practices
Abstract
A system and methods for generating a profile concerning the
prescription therapy practices of physicians in a therapeutic area
of interest. A plurality of historical de-identified patient
prescription records are analyzed, and the records are categorized
based on changes in the dosage or the prescription product between
prescriptions for a particular de-identified patient. The
prescriptions may be categorized based on the duration between the
due date of a prescription and the date the prescription was
filled. This information may be summarized for each physician for
each product in the therapy area of interest.
Inventors: |
Tolle, Jill; (Collegeville,
PA) ; Cain, James; (Doylestown, PA) |
Correspondence
Address: |
BAKER & BOTTS
30 ROCKEFELLER PLAZA
NEW YORK
NY
10112
|
Family ID: |
26956387 |
Appl. No.: |
09/941496 |
Filed: |
August 29, 2001 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
60273714 |
Mar 5, 2001 |
|
|
|
Current U.S.
Class: |
705/3 |
Current CPC
Class: |
G06Q 10/10 20130101;
G16H 15/00 20180101; G16H 70/40 20180101; G16H 10/20 20180101; G16H
20/10 20180101 |
Class at
Publication: |
705/3 |
International
Class: |
G06F 017/60 |
Claims
We claim:
1. A method for generating a profile concerning prescription
therapy practices of at least one physician in a therapeutic area
of interest, comprising the steps of: (a) receiving a plurality of
historical de-identified patient prescription records corresponding
to prescriptions issued to at least one de-identified patient by at
least one physician, each record including de-identified patient
identification number, dosage, and prescription product
information; (b) receiving user-specified information defining a
subset of the historical de-identified patient prescription
records; (c) extracting at least one relevant historical
de-identified patient prescription record from the received
historical de-identified patient prescription records based on the
subset; (d) for each de-identified patient, comparing dosage and
prescription product information contained in a first extracted
historical de-identified patient record with dosage and
prescription product information contained in a second extracted
historical de-identified patient record; and (e) for each
comparison made in step (d), categorizing a prescription based on a
change in dosage or prescription product.
2. The method of claim 1, wherein the step of receiving
user-specified information further comprises receiving information
concerning an observation period.
3. The method of claim 1, wherein the step of extracting
de-identified patient prescription records further comprises
discarding a historical de-identified patient prescription record
having an eligibility of shorter duration than the observation
period.
4. The method of claim 1, wherein the step of categorizing
prescriptions further comprises categorizing a prescription of a
product as a new therapy start when a de-identified patient has had
no other prescriptions in a therapeutic area to which the product
pertains.
5. The method of claim 1, wherein the step of categorizing
prescriptions further comprises categorizing a prescription as a
therapy switch when a de-identified patient has had no other
prescriptions of the product, and when the de-identified patient
has had a prescription of a second product within a therapeutic
area to which the product pertains that was not prescribed on the
same day as the prescription.
6. The method of claim 1, wherein step of categorizing
prescriptions further comprises categorizing a prescription as an
add-on therapy when the de-identified patient had no other
prescriptions for the product but had a prescription for a second
product within the therapeutic area to which the product pertains
that was prescribed on the same day.
7. The method of claim 1, wherein the step of categorizing
prescriptions further comprises categorizing a prescription as a
titration decrease when a de-identified patient has previously had
a prescription for the product at a higher dosage.
8. The method of claim 1, wherein the step of categorizing
prescriptions further comprises categorizing a prescription as a
titration increase when a de-identified patient has previously had
a prescription for the product at a lower dosage.
9. The method of claim 1, wherein the step of categorizing
prescriptions further comprises categorizing a prescription as
continued therapy when a de-identified patient has previously had a
prescription for the product at the same dosage.
10. A method for generating a profile concerning prescription
therapy practices of at least one physician in a therapeutic area
of interest comprising the steps of: (a) receiving a plurality of
historical de-identified patient prescription records corresponding
to prescriptions issued to at least one de-identified patient by at
least one physician, each record including de-identified patient
identification number, prescription product information, date
dispensed, dosage, number of days supplied, and refill information;
(b) receiving user-specified information defining a subset of the
historical de-identified patient prescription records; (c)
extracting at least one relevant historical de-identified patient
prescription record from the received historical de-identified
patient prescription records based on the subset; (d) for each
de-identified patient, comparing dosage and prescription product
information contained in a first extracted historical de-identified
patient record with dosage and prescription product information
contained in a second extracted historical de-identified patient
record; (e) for each comparison made in step (d), categorizing a
prescription based on a change in dosage or prescription product;
(f) extracting at least one relevant historical de-identified
patient prescription record from the prescriptions categorized at
step (e) based on the refill information; (g) for each
de-identified patient, determining a refill due date based on the
dosage and the number of days supplied for a first prescription;
(h) for each de-identified patient, comparing the refill due date
of the first prescription with the date dispensed for a second
prescription; and (i) for each comparison made in step (h),
categorizing the de-identified patient based on the duration
between the refill due date of the first prescription and the date
dispensed for the second prescription.
11. The method of claim 10, wherein the step of receiving
user-specified information further comprises receiving information
concerning an observation period.
12. The method of claim 11, further comprising a step of
discarding, after step (e), a historical de-identified patient
prescription record for a de-identified patient having more than
one physician during the observation period.
13. The method of claim 11, wherein the step of extracting
de-identified patient prescription records based on refill
information further comprises discarding a historical de-identified
patient prescription record not having a refill due within the
observation period.
14. The method of claim 10, wherein the step of categorizing
prescriptions further comprises categorizing a prescription of a
product as a new therapy start when a de-identified patient has had
no other prescriptions in a therapeutic area to which the product
pertains.
15. The method of claim 10, wherein the step of categorizing
prescriptions further comprises categorizing a prescription as a
therapy switch when a de-identified patient has had no other
prescriptions of the product, and when the de-identified patient
has had a prescription of a second product within a therapeutic
area to which the product pertains that was not prescribed on the
same day as the prescription.
16. The method of claim 15, further comprising a step of
discarding, after step (e), a historical de-identified patient
prescription record for a prescription categorized as a therapy
switch.
17. The method of claim 10, wherein the step of categorizing
prescriptions further comprises categorizing a prescription as a
titration decrease when a de-identified patient has previously had
a prescription for the product at a higher dosage.
18. The method of claim 17, further comprising a step of
discarding, after step (e), a historical de-identified patient
prescription record for a prescription categorized as a titration
decrease.
19. The method of claim 10, wherein the step of categorizing
prescriptions further comprises categorizing a prescription as a
titration increase when a de-identified patient has previously had
a prescription for the product at a lower dosage.
20. The method of claim 19, further comprising a step of
discarding, after step (e), a historical de-identified patient
prescription record for a prescription categorized as a titration
increase
21. The method of claim 10, wherein the step of categorizing the
de-identified patient based on the duration between the due date of
the first prescription and the fill date of the second prescription
further comprises categorizing the de-identified patient as
persistent if the duration is shorter than a predetermined number
of days.
22. The method of claim 21, wherein the step of categorizing the
de-identified patient based on the duration between the due date of
the first prescription and the fill date of the second prescription
further comprises categorizing the de-identified patient as
non-persistent if the duration is greater than a predetermined
number of days.
23. The method of claim 22, further comprising a step of
determining, after step (i), the total number of persistent
de-identified patients and the total number of non-persistent
de-identified patients for each physician.
24. The method of claim 23, further comprising a step of
calculating, after the step of determining the total number of
persistent de-identified patients and the total number of
non-persistent de-identified patients for each physician,
persistence of the physician by dividing the total number of
persistent de-identified patients by the total number of
de-identified patients for each physician.
25. A system for generating a profile concerning prescription
therapy practices of at least one physician in a therapeutic area
of interest, comprising: (a) a mass storage device for storing a
plurality of historical de-identified patient prescription records
corresponding to prescriptions issued to at least one de-identified
patient by at least one physician, each record including a
de-identified patient identification number, dosage, number of days
supplied and prescription product information, dosage, fill date,
and number of days supplied; (b) an input device, coupled to the
mass storage device, for receiving user-specified information which
defines a subset of the plurality of historical de-identified
patient prescription records; (c) a prescription categorizer,
coupled to the input device, configured to compare the dosage and
the prescription product information contained in a first
historical de-identified patient prescription record with the
dosage and prescription product information contained in a second
historical de-identified patient prescription record, and to
categorize a prescription based on a change in dosage or
prescription product; and (d) a persistence calculator, coupled to
the prescription categorizer, configured to determine the due date
of a first prescription based on the dosage and the number of days
supplied, to compare the due date of the first prescription with
the fill date of a second prescription, and to categorize the
de-identified patient based on the duration between the due date of
the first prescription and the fill date of the second
prescription.
26. The system of claim 25, wherein the prescription categorizing
is configured to categorize a prescription of a product as a new
therapy start when a de-identified patient has had no other
prescriptions in a therapeutic area to which the product
pertains.
27. The system of claim 25, wherein the prescription categorizing
is configured to categorize a prescription of a product as a
therapy switch when a de-identified patient has had no other
prescriptions of the product, and when the de-identified patient
has had a prescription of a second product within a therapeutic
area to which the product pertains that was not prescribed on the
same day as the prescription.
28. The system of claim 25, wherein the prescription categorizing
is configured to categorize a prescription of a product as an
add-on therapy when the de-identified patient had no other
prescriptions for the product but had a prescription for a second
product within the therapeutic area to which the product pertains
that was prescribed on the same day.
29. The system of claim 25, wherein the prescription categorizing
is configured to categorize a prescription of a product as a
titration decrease when a de-identified patient has previously had
a prescription for the product at a higher dosage.
30. The system of claim 25, wherein the prescription categorizing
is configured to categorize a prescription of a product as a
titration increase when a de-identified patient has previously had
a prescription for the product at a lower dosage.
31. The system of claim 25, wherein the prescription categorizing
is configured to categorize a prescription of a product as
continued therapy when a de-identified patient has previously had a
prescription for the product at the same dosage.
32. The system of claim 25, wherein the persistence calculator is
further configured to categorize the de-identified patient as
persistent if the duration between the due date of the first
prescription and the fill date of the second prescription is
shorter than a predetermined number of days.
33. The system of claim 25, wherein the persistence calculator is
further configured to categorize the de-identified patient as
non-persistent if the duration between the due date of the first
prescription and the fill date of the second prescription is
greater than a predetermined number of days.
34. The system of claim 25, wherein the persistence calculator is
further configured to determine the total number of persistent
de-identified patients and the total number of non-persistent
de-identified patients for each physician.
35. The system of claim 34, wherein the persistence calculator is
further configured to determine the persistence of the physician by
dividing the total number of persistent de-identified patients by
the total number of de-identified patients for each physician.
Description
CROSS-REFERENCE TO RELATED APPLICATION
[0001] This application claims priority on U.S. Provisional Patent
Application Serial No. 60/273,714, entitled "Physician Profiling
for Enhanced Sales Force Targeting Using Longitudinal Rx Data
(LRx)," which was filed on Mar. 5, 2001, and which is incorporated
by reference in its entirety herein.
BACKGROUND OF THE INVENTION
[0002] 1. Field of the Invention
[0003] The invention relates to systems and methods for analyzing
prescription claim histories for physicians, and determining
statistics about the prescription therapies of such physicians.
[0004] 2. Related Art
[0005] Pharmaceutical sales representatives typically determine a
territory call plan based on information about physicians in their
respective coverage areas, and the range of pharmaceutical products
that such physicians typically prescribe. This information may
include the specialty of the physician, the physician's response to
promotional efforts, the physician's ranking in the pharmaceutical
product's market share, the physician's ranking in total market
volume, and the physician's ranking in the pharmaceutical product's
prescription volume. Based on observed patterns with respect to
this information, further qualities about physicians have been
successfully modeled such as "new product early adopter," which
refers to physician who tends to prescribe a new product soon after
it becomes available, or "brand loyalist," which refers to a
physician who continues to prescribe a specific branded drug, even
in the face of competitive drug availability.
[0006] While the above information is derived from prescriptions
written by the physicians, the information does not provide insight
into either physicians treatment practices over a given period of
time, or such practices as applied to different patient types. Such
targeting would require a more detailed understanding of a
physician's treatment practices within his patient population,
e.g., through the formation of a database of prescription activity
for each physician where de-identified patients can be tracked to
understand how a physician prescribes in a particular therapeutic
area.
[0007] Although algorithms have been developed for use on
longitudinal medical records, these applications have been for
market research or summarized levels of analysis. Outcome research
and market research on longitudinal prescription data is currently
being offered through several vendors - all utilizing single
sources of data for their applications.
[0008] Accordingly, there exists a need in the art for a technique
which can analyze the long term prescribing practices of
physicians, including the ability to provide information about
changes in both dosages and products prescribed by the physicians
and the ability to provide information about the tendencies of a
physician's patient population to follow prescribed regimens.
SUMMARY OF THE INVENTION
[0009] An object of the present invention is to provide a technique
for analyzing the prescription practices of multiple physicians
over a given period of time.
[0010] Another object of the present invention is to provide
prescription activity analysis tools which can assist
pharmaceutical sales representatives in understanding the
prescription practices of physicians.
[0011] A further object of the present invention is to provide a
technique which can track changes in prescription dosages
prescribed by a physician.
[0012] A still further object of the present invention is to
provide a technique for monitoring patient population tendencies to
follow prescriptions regimens.
[0013] These and other objects of the invention, which will become
apparent with reference to the disclosure herein, are accomplished
by a system and method for generating a longitudinal physician
profile concerning the prescription therapies for de-identified
patients issued by one or more physicians in a particular
therapeutic area of interest. A physician profile may be generated
on a computer system including a mass storage device containing
de-identified historical patient prescription records categorized
by de-identified patient, physician, prescription product and
dosage.
[0014] The historical de-identified patient prescription records
may be retrieved from the mass storage device by a dynamic memory,
and user-specified information may be provided to define a subset
of the historical de-identified patient prescription records.
Particular historical prescription data within the subset of data
defined by the user-specified information may be extracted to
provide a set of data including de-identified patients that meet
certain eligibility requirements.
[0015] In one preferred arrangement, the categorization process of
each prescription may include comparing the dosage and prescription
product for a particular prescription for each patient with the
dosage and prescription product of another prescription for the
patient. The categorization of the particular prescription is based
on a change in the dosage or in the prescription product between
the particular prescription and the other prescription. Particular
categories of prescriptions may include a new therapy start, a
therapy switch, an add-on therapy, a titration increase or
decrease, or a continued therapy.
[0016] Advantageously, persistence may be determined both for each
particular patient, as well as for all patients for each physician.
The patient prescription data may be extracted with a second subset
of data defined by a categorizing step. For example, prescriptions
that are a therapy switch or a titration may be excluded from the
compliance/persistence analysis. The duration between a due date of
a particular prescription and the date the prescription was filled
may be determined. The particular prescription may be categorized
based on the duration between the due date and the date the
prescription was filled.
[0017] In accordance with the invention, the objects as described
above have been met, and the need in the art for a technique which
can analyze the long term prescription practices of a group of
physicians, including the ability to provide information about
changes in both dosages and products prescribed by the physicians
and the ability to provide information about the tendencies of a
physician's patient population to follow prescribed regimens, has
been satisfied. Further features of the invention, its nature and
various advantages will be more apparent from the accompanying
drawings and the following detailed description of illustrative
embodiments.
BRIEF DESCRIPTION OF THE DRAWINGS
[0018] FIG. 1 is a block diagram of an exemplary system in
accordance with the invention.
[0019] FIG. 2 is a flowchart illustrating an exemplary procedure in
accordance with the invention.
[0020] FIG. 3 is a flowchart illustrating a portion of the
procedure of FIG. 2 in greater detail in accordance with the
invention.
[0021] FIG. 4 is an exemplary summary report in accordance with the
invention.
[0022] FIG. 5 is flowchart illustrating another portion of the
procedure of FIG. 2 in greater detail in accordance with the
invention.
[0023] FIG. 6 is a flowchart illustrating a portion of the
procedure of FIG. 5 in greater detail in accordance with the
invention.
[0024] FIG. 7 is a flowchart illustrating another portion of the
procedure of FIG. 5 in greater detail in accordance with the
invention.
[0025] FIG. 8 is a flowchart illustrating a further portion of the
procedure of FIG. 5 in greater detail in accordance with the
invention
[0026] FIG. 9 is a summary of physician prescription
information.
[0027] FIG. 10 is an exemplary summary report in accordance with
the invention.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0028] Referring now to FIG. 1, an illustrative embodiment of a
system for processing prescription data is depicted and generally
referred to as system 10. The system 10 may utilize several sources
of information for processing. The user supplies information on a
particular therapeutic area or market of interest 12, such as an
anti-depressant therapy or blood pressure control therapy. The user
may also supply information on certain prescription products which
are to be included in the study 14. Time period information 15,
i.e., an "observation period," is selected by the user to specify
the period of time in which to monitor the dispensing of
prescriptions. Information on the specific prescriptions is
included in prescription data, i.e., Retail Pharmacy Prescription
data 16, which includes historical de-identified patient
prescription data and is typically stored on a mass storage device.
This input information in received by the system at the input
device 18.
[0029] The physician targeting system 10 uses longitudinal
prescription data from retail pharmacies. Retail Pharmacy
Prescription data 16 supplies information such as the prescription
product dispensed, the dosage, refill information, i.e., an
indication of whether or not a refill is authorized, the day
supply, i.e., the number of days until the patient will need a
refill, and the date dispensed. Retail Pharmacy Prescription data
16 groups the above information for one patient under a
"de-identified" patient identification number. The de-identified
patient identification number is an identifier that replaces a
patient's name and protects patient confidentiality since it
provides no personal information about the patient. This
information allows the system to track prescription therapy over
time for one specific, although unknown, patient. Thus whenever a
"patient" or "patient data" is described herein, it is understood
that the patient's identity and personal information are excluded
(i.e., the patient is "de-identified") in order to maintain
confidentiality of patient records.
[0030] While the disclosure herein is described with use of Retail
Pharmacy Prescription data, other data structures could readily be
employed, such as Pharmacy Benefit Manager (PBM) prescription
claims data, mail order prescription data, or a combination of data
sources.
[0031] A prescription categorizer 20 and compliance and persistence
calculator 22 perform a series of data processing operations by the
central processing unit of a computer, executing software programs
in COBOL, which are stored in dynamic computer memory, such as RAM
(not shown). The input data is received by the prescription
categorizer 20 which first considers whether each de-identified
patient is "track-able" to be included into the prescription
categorization process. Once track-ability is confirmed, then the
prescription categorizer 20 compares the dosage and prescription
product for a particular prescription for a each de-identified
patient with the dosage and prescription product of another
prescription for that de-identified patient identification number
and categorizes the particular prescription based on a change in
the dosage or the prescribed medication between the particular
prescription and the other prescription. Each prescription may be
categorized by the prescription categorizer 20 into the following
categories: (1) New Therapy Start, (2) Therapy Switch, (3) Add-on
Therapy (concomitant), (4) Titration Decrease, (5) Titration
Increase, or (6) Continued Therapy. Those skilled in the art will
understand that other categories could be added.
[0032] The persistence calculator 22 compares the due date of a
particular prescription for a de-identified patient with the date
the subsequent prescription was actually filled. The persistence
calculator 22 also categorizes the de-identified patient based on
the number of days between the due date and the date the
prescription was filled. The above calculations performed by the
software program. Persistence in the prescription data refers to a
de-identified patient's diligence in following their physician's
prescription therapy. The persistence calculator may also summarize
the categories of de-identified patients for each physician.
Compliance is a measure attributed to the physician derived from
the percentage of days that persistent de-identified patients were
on a prescription therapy when compared with the number of days
that the de-identified patients were supposed to be on that
therapy. These two measures are determined by the compliance and
persistence calculator 22. This information is organized into a
summary which is stored as output data 26. This is described in
greater detail as noted below with respect to FIGS. 9-10.
[0033] The procedures implemented by the present invention are
described with respect to FIGS. 2-3 and 5-8. The foregoing
procedures may be implemented on a general purpose computer (not
shown) including a suitable input device 18 and mass storage 26.
With specific reference to FIG. 2, information on the therapeutic
area of interest, the prescription products included in the study,
and the observation period is received at step 100. The Retail
Pharmacy Prescription data for the transactions to be included in
the study are received at step 102. Note that these steps may
proceed consecutively or concurrently. Subsequently, each of the
prescriptions are categorized into one of several predetermined
categories, as listed above, in step 104. Compliance and
persistence information is determined at step 106. A summary of
information determined in steps 104 and 106 is generated and stored
at step 108.
[0034] Further details of step 104 are illustrated in FIG. 3. In
order for a de-identified patient's prescription to be included
into the longitudinal prescription categorization process, patient
eligibility files are used to determine if a de-identified patient
has been prescription benefit eligible for a specified number of
weeks, i.e., the observation period. This minimum period of time is
useful to get a full view of a de-identified patient's prescription
therapy history. For example, a de-identified patient who is newly
enrolled in a health plan would not have sufficient historical
information to determine whether a prescription was a continuation
or a change in a particular therapy. Thus, if the de-identified
patient is not considered eligible, the data for that de-identified
patient is discarded.
[0035] Information on the therapeutic area, the prescription
products of interest, and observation period received in step 100
and Retail Pharmacy Prescription data received in step 102 are
passed on to step 120 where a patient eligibility extract is
generated. A filter is used to create a list of de-identified
patient identification numbers for patients that are eligible for
the prescription plan during the observation period. The patient
eligibility extract may also be limited to the specified products
in the specified market of interest as defined by the user.
[0036] Steps 122-146 may be performed for each prescription in the
analysis. The prescription being categorized in steps 122-146 is
referred to herein as the "current prescription." The process of
categorization for the current prescription begins at step 122. In
accordance with a preferred embodiment of the invention, it is
determined whether the current prescription is a "New Therapy
Start" for the de-identified patient at step 122. A New Therapy
Start may be defined as the first prescription that a de-identified
patient receives in the market of interest. This is performed by
searching the records to see if the de-identified patient has any
earlier prescriptions in the market of interest during the
observation period. If there were no other prescriptions in this
market for that de-identified patient and the current prescription
has a "new prescription" indicator in the Retail Pharmacy
Prescription data, then the current prescription is categorized as
a New Therapy Start (step 124).
[0037] At step 126, a prescription that has been categorized as
"New Therapy Start" is compared to the patient eligibility extract.
If the de-identified patient identification number on the current
prescription is not listed in the patient eligibility extract, then
the current prescription is discarded from the categorization
process for this de-identified patient at step 128. The process
continues with the next prescription as indicated by marker A in
FIG. 3
[0038] If there was at least one other "market prescription,"
defined as a previous prescription in the market of interest during
the observation period, then the analysis proceeds to step 130.
This step determines whether that other market prescription was for
the same product as the current prescription. If the other market
prescription was for a different product, the analysis proceeds to
step 132. There are two preferable categories when the other market
prescription was for a different product: (1) a "Therapy Switch"
which occurs when a de-identified patient receives a prescription
for a different product than the last one they received, and (2) an
"Add-on Therapy" occurs when a de-identified patient appears to be
taking more than one product at a time, i.e., the prescription is
"added-on" to other market prescriptions, as will be described in
greater detail herein.
[0039] At step 132, the dates of the current prescription and the
other market prescription are considered. If the other market
prescription was not dispensed on the same day as the current
prescription, then the current prescription is categorized as a
Therapy Switch at step 134 and the process is finished for the
current prescription. The current prescription may also store
information on the product it was switched from. However, if the
current prescription and the other market prescription were
dispensed on the same date, then the current prescription is
considered to be an Add-On Therapy at step 136, and the process is
finished for the current prescription. After steps 134 and 136, the
process continues with the next prescription as indicated by marker
A. Although the embodiment described herein defines a prescription
having the same date as an Add-On therapy, the system may be
configured to permit a date range or window that would further
define the difference between a therapy switch and an add-on.
[0040] If the step 130 determined that the current prescription and
the other market prescription are for the same product, then
analysis proceeds to step 138. There are three preferable
categories for a current prescription which is for the same product
as the other market prescription in the records: (1) A Continued
therapy is a prescription at the same dosage level as the
de-identified patient's last prescription; (2) A Titration Increase
is a prescription for the same product at a greater dosage level
than that de-identified patient's last prescription; and (3) A
Titration Decrease is a prescription for the same product at a
smaller dosage level. Other categories of prescriptions could also
be utilized.
[0041] At step 138, the daily dosages for the current prescription
and the other market prescription are compared. Dosage is computed
as the strength of the product multiplied by the amount of product
to be taken each day. For example, product Y may have a strength of
50 mg and require 2 pills per day. The daily dosage for product Y
is therefore 100 mg (i.e., 50 mg * 2 pills). Other measures of
dosage rate may be utilized as appropriate. If the dosages for the
current prescription and the other market prescription are the
same, or if either of the dosages is invalid then the current
prescription is categorized as a Continued Therapy at step 140, and
the process is finished for the current prescription. At step 142,
if the dosage on the other market prescription is found to be lower
than the dosage of the current prescription, the current
prescription is categorized as a Titration Increase at step 144,
and the process is finished for the current prescription. If the
dosage for the other market prescription is not lower than the
dosage for the current prescription, i.e., the other market
prescription was greater than the current prescription, then the
current prescription is categorized as a Titration Decrease at step
146, and the process is finished for the current prescription. If
there is another prescription, as determined at step 148, the
categorization process continues for that prescription, as
indicated by marker A. When the last prescription is reached, the
process proceeds to step 160, as described below.
[0042] With the process described above, each prescription for a
track-able de-identified patient will preferably fall into one
category. It is noted that if additional prescription categories
are utilized, each prescription may fall into more than one such
category. Of course, the sequence of steps described above with
respect to steps 120-146 may be changed in order to determine the
category of a prescription. Finally, prescription category counts
may be added and summarized for each physician or product for any
desired time period.
[0043] Example
[0044] FIG. 4 provides an example of prescription categorization.
In this example, therapeutic area, or market of interest, is the
anti-depressant therapy market. The exemplary time frame 300
includes weekly views from the period beginning with the first
quarter of 1999 to the third quarter of 1999. The exemplary
observation window consists of the 13 weeks prior to any reporting
week. Since these prescriptions all contain the same de-identified
patient number, they are categorized according to the rules
described above with respect to FIG. 3. The particular prescription
number, i.e., Prescription #1 to Prescription #16, appears in
column 301. Column 302 indicates the product prescribed, column 304
indicates the dosage, and column 306 has the start date
information. Each entry is indicated by a triangle 308 representing
the first date of the prescription, and a triangle 310 indicating
the last date of the prescription. In this example, Prescription #1
is a New Therapy start. There were no other prescriptions in this
market for this de-identified patient number in the prior 13 weeks
before Feb. 7, 1999. Based on the cess described above, the
remaining prescriptions are categorized as shown in TABLE 1
1TABLE 1 Prescription Category Comments 1 New Therapy start No
previous market prescription 2 Continuing Therapy Same product,
same dosage 3 Titration Increase 40 mg increased from 20 mg 4
Therapy Switch Drug #1 to Drug #2 5 Titration Increase 225 mg
increased from 150 mg 6 Continuing Therapy Same product, same
dosage 7 Continuing Therapy Same product, same dosage 8 Titration
Increase 300 mg increased from 225 mg 9 Therapy Switch Drug #2 to
Drug #3 10a Therapy Switch Drug #3 to Drug #2 10b Add-on Therapy
Drug #2 (75 mg) same day as Prescription 10a 10c Add-on Therapy
Drug #4 dispensed same day as Prescription 10a 11a Continuing
Therapy Same product, same dosage as Prescription 10a 11b
Continuing Therapy Same product, same dosage as Prescription 10b
11c Continuing Therapy Same product, same dosage as Prescription
10c 12 Titration Increase 30 mg increased from 15 mg 13a Continuing
Therapy Same product, same dosage as Prescription 11b 13b
Continuing Therapy Same product, same dosage as Prescription 11a 14
Continuing Therapy Same product, same dosage as Prescription 12 15
Continuing Therapy Same product, same dosage as Prescription 13a 16
Continuing Therapy Same product, same dosage as Prescription
13b
[0045] Referring again to FIG. 4, the three types of arrows
represent the three unique physicians prescribing during this
course of treatment, i.e., Physician A in dashed outline, Physician
B in solid outline, and Physician C in solid color. At the end of
the categorization, the prescriptions may be aggregated to the
physician, product, category level for each weekly reporting time
period.
[0046] FIG. 5 provides further details of step 106 of calculating
therapy persistence and compliance. As described above, persistence
and compliance are measures of a de-identified patient's diligence
in following their doctor's prescription therapy. The time period
for evaluation was selected by the user at step 100 (See FIG. 2).
In order to perform the persistence evaluation, the analysis
according to the invention preferably uses one additional month of
data beyond the defined measurement time period at step 160 of FIG.
5, although other time periods could be utilized. For example, to
report compliance and persistence for a time period (month or
quarter) ending in January, then February data is used. Selecting
de-identified patients who continued therapy is determined at step
160. Persistence data is determined for each de-identified patient
at step 162. A de-identified patient is labeled "persistent" if
they refilled their prescription on time. If they were late in
refilling a prescription they are considered "non-persistent." A
further stage of the process is to calculate the persistence and
compliance data for each physician at step 164. As will be
described in greater detail herein, compliance is a physician
measure based on the percentage of days that their persistent
de-identified patients were supposed to be on a drug therapy
divided by the total number of days those de-identified patients
were actually on that therapy. These applications are suitable on a
weekly, monthly or quarterly basis.
[0047] FIG. 6 provides further details with respect to step 160.
Step 170 is to determine whether de-identified patients that have
at least one market prescription due for refill within the selected
time period. If the de-identified patient does not have a
prescription due for refill within the selected time period, the
data is discarded at step 172, and the next de-identified patient
data is evaluated at step 170. In order to calculate the refill due
date, the day supply is added to the prescription dispense date on
the prescription. Using this due date, step 170 selects all
de-identified patients who have a refill due in the observation
period. For example, if a January observation period has been
chosen, all de-identified patients who have a prescription with a
refill due date within the month of January are selected. Patient
data is discarded at step 172 for de-identified patients who do not
have prescription with a refill due date during the time
period.
[0048] The next step, step 174, is to select all prescriptions for
the de-identified patients remaining from step 170 during of the
time period and extending to a follow-up period. For example, for a
monthly report, the observation period may be the month of January
and the follow-up period is the month of February. Using the
prescription categories previously determined in steps 126-146 as
described above with respect to FIG. 3, steps 178-184 discard data
for any de-identified patient who had either a switch or a
titration in therapy. More particularly, at step 178, if the
prescription category is determined to be a Therapy Switch, then
the data is discarded at step 172 (as illustrated by Marker C). If
the prescription is not a Therapy Switch, then the process
continues to step 180, in which the data is discarded if the
prescription is a Titration Increase. If the prescription is not a
Titration Increase, the prescription category is determined at step
182. Here the data is discarded if the prescription is a Titration
Decrease. Otherwise, the process determines whether the
de-identified patient had more than one physician prescribing
prescriptions at step 184. Data is discarded for de-identified
patients who had more than one physician at step 172. The remaining
data is for de-identified patients having continuing therapy with
one doctor, which proceeds to step 190 (FIG. 7).
[0049] Further details of step 162, i.e., calculating the
persistence data for each de-identified patient is illustrated in
FIG. 7. A stage in the process is to compare the date of the last
prescription within the observation period (e.g., January) and the
date of a subsequent prescription for that product in the follow-up
period (e.g., February). More particularly, the date of a
prescription in the follow-up period is determined at step 190. If
it is determined at step 192 that the prescription was filled less
than 30 days after the due date, that de-identified patient is
labeled "persistent" at step 194. Otherwise, the de-identified
patient is labeled "non-persistent" at step 196. Other time periods
could readily be utilized, as could a multi-tiered persistence
categorization.
[0050] For the persistent de-identified patients, the total number
of days supplied for each de-identified patient is determined at
step 198. In the embodiment described herein, this data may be
obtained from the Retail Pharmacy Prescription data. The total
number of days on therapy is determined at step 200. More
particularly, the number days on therapy may be calculated by
counting the number of days between the dispensed date on the
initial prescription in the observation period (e.g., the January
prescription) and the dispensed date of the subsequent prescription
during the follow-up period (e.g., the February prescription). At
step 202, if the next de-identified patient data is to be analyzed,
the analysis returns to step 170 for the next de-identified
patient, as described with respect to FIG. 6. If there is no
additional patient data, the analysis proceeds to step 204 (FIG.
8).
[0051] In FIG. 8, further details are described with respect to the
calculation of compliance and persistence data for a physician
(i.e., step 164 of FIG. 5). At step 204, the total number of
non-persistent de-identified patients for a physician is
calculated. The total number of persistent de-identified patients
is determined at step 206. Next, the total days of therapy for all
persistent de-identified patients for this physician are calculated
at step 208. The total days supply for all persistent de-identified
patients for this physician are calculated at step 210. In the
following steps 204, 206, 208, and 210, data for the total number
of persistent de-identified patients, the total number of
non-persistent de-identified patients, the total number of
de-identified patients, i.e., the total number of persistent and
non-persistent de-identified patients, the total days supply for
persistent de-identified patients, and the total days of therapy
for persistent de-identified patients are merged into a single
doctor/product level file at step 212.
[0052] At step 214, the compliance data for the physician is
computed as follows:
Compliance=total days supply/total days on therapy (1)
[0053] Persistence data for the physician/product level is
calculated at step 216:
Persistence=Number of persistent patients/Total number of patients
(2)
[0054] As illustrated in FIG. 2, a summary report for each
physician is generated at step 108. After the prescriptions are
categorized and aggregated to a physician/product/category level,
the data is summarized in a report. An example of a summarized
record 350 is illustrated in FIG. 9, and stored as output data 26.
The level of summarization is to the physician and prescription
product. As described above with respect to steps 122-146 of FIG.
3, the categorization of prescriptions occurs at the individual
prescription level. Within the time period of interest, the
categorized transactions are summed up in each of the categories.
In summarized record 350, the data has been summarized for
physician 352 and for prescription product 354. The record 350
therefore provides the total number of new therapy starts 356,
continued therapy 358, therapy switches 360, titration increases
362 and titration decreases 364 for physician 352 and prescription
product 354. The persistence and compliance data calculated in
steps 160, 162, and 164 FIG. 5) are also summarized in report 350.
More particularly, summarized record 350 contains data for the
persistence 366 and compliance 368 for each physician 352 and
prescription product 364. Filters may be applied to enable the
system to extract only the information that is actionable by sales
representatives. Such filters can be based upon either analytical
procedures or business rules or a combination of both. For example,
a filter is used to select the top 20 doctors in a territory based
on market prescription volume.
[0055] In practice, a sales representative may only want to be
notified when doctors in his territory start a statistically
significant number of patients on a competitor's product for second
line therapy. Armed with this knowledge, the sales representative
may use a very specific selling message to inform these doctors
about the clinical benefits of using the company's brand for second
line therapy. Thus, the system described herein could be adapted to
extract and deliver exception-based alerts and cycle appropriate
information insight to the sales representatives with the right
message at the right time.
[0056] FIG. 10 illustrates an exemplary report 400 that may be
delivered to a sales representative. The territory level report
reformat is generated by software code in a back-end reporting
system. The system accepts parameters that control what data and
how much data are sent to the report. The system customizes the
report to the customer's needs based on the parameters selected.
Exemplary report 400 has been selected to display statistics for
products DRUG #1 and DRUG #2 in the "chronic pain" market of
interest 402. Other parameters selected are "sales force" 404 and
"territory" 406. A summary portion 410 of the report 400 contains
the reported products' shares 414 and market volume 416 by week,
which is currently available in today's environment. The report 400
may also provide information about physician practice in a
prescriber portion 412. For example, prescriber portion 412 of FIG.
10 illustrates the top four physicians in sections 420a, 420b,
420c, and 420d. In addition, a text message 422a, 422b, 422c, 422d,
further describing an observed trend in prescribing practice, is
attached to each physician. Statistically significant trends based
on the categorized prescription data are determined by applying
statistical modeling and trending tests. Each physician is given a
score from the statistical trend tests. If the score falls within a
user defined range, then that physician is eligible for an alert.
For example, Dr. John Doe, once ranked at #21 on market volume, is
now ranked at #1 based on longitudinal prescription activity. A
sales representative viewing this report now has a clearer picture
of Dr. Doe's behavior and can tailor his message appropriately and
quickly.
[0057] One skilled in the art will appreciate that the present
invention can be practiced by other than the described embodiments,
which are presented here for purposes of illustration and not of
limitation, and the present invention is limited only by the claims
that follow.
* * * * *